{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:PX22RPVBQX5BOAKLCJO5ECX4E6","short_pith_number":"pith:PX22RPVB","schema_version":"1.0","canonical_sha256":"7df5a8bea185fa17014b125dd20afc279d71b45142f712eb699df0391cc41bd1","source":{"kind":"arxiv","id":"2005.03687","version":2},"attestation_state":"computed","paper":{"title":"COBRA: Contrastive Bi-Modal Representation Algorithm","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhishek Maiti, Deepak Srivatsav, Rajiv Ratn Shah, Suryatej Reddy Vyalla, Vishaal Udandarao, Yifang Yin","submitted_at":"2020-05-07T18:20:12Z","abstract_excerpt":"There are a wide range of applications that involve multi-modal data, such as cross-modal retrieval, visual question-answering, and image captioning. Such applications are primarily dependent on aligned distributions of the different constituent modalities. Existing approaches generate latent embeddings for each modality in a joint fashion by representing them in a common manifold. However these joint embedding spaces fail to sufficiently reduce the modality gap, which affects the performance in downstream tasks. We hypothesize that these embeddings retain the intra-class relationships but are"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2005.03687","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-05-07T18:20:12Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"9c2514e63042f2a0129d390445f3038248093a79fd7c1dd8f7f668804a580a04","abstract_canon_sha256":"0e9e015c59a033d8f014bcfd84c815587d96ff8fceb18a89e4a98220284bc8a4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:05:27.347491Z","signature_b64":"RATBsotkRb5T8PBHadH/LVeboGSRLtANyBV+TAD0ja70NbQfyJFBnGXuEUZoN2h/PnCnKpMIWNxpXSUp1E8RBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7df5a8bea185fa17014b125dd20afc279d71b45142f712eb699df0391cc41bd1","last_reissued_at":"2026-07-05T01:05:27.346993Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:05:27.346993Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"COBRA: Contrastive Bi-Modal Representation Algorithm","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhishek Maiti, Deepak Srivatsav, Rajiv Ratn Shah, Suryatej Reddy Vyalla, Vishaal Udandarao, Yifang Yin","submitted_at":"2020-05-07T18:20:12Z","abstract_excerpt":"There are a wide range of applications that involve multi-modal data, such as cross-modal retrieval, visual question-answering, and image captioning. Such applications are primarily dependent on aligned distributions of the different constituent modalities. Existing approaches generate latent embeddings for each modality in a joint fashion by representing them in a common manifold. However these joint embedding spaces fail to sufficiently reduce the modality gap, which affects the performance in downstream tasks. We hypothesize that these embeddings retain the intra-class relationships but are"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.03687","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2005.03687/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2005.03687","created_at":"2026-07-05T01:05:27.347053+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.03687v2","created_at":"2026-07-05T01:05:27.347053+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.03687","created_at":"2026-07-05T01:05:27.347053+00:00"},{"alias_kind":"pith_short_12","alias_value":"PX22RPVBQX5B","created_at":"2026-07-05T01:05:27.347053+00:00"},{"alias_kind":"pith_short_16","alias_value":"PX22RPVBQX5BOAKL","created_at":"2026-07-05T01:05:27.347053+00:00"},{"alias_kind":"pith_short_8","alias_value":"PX22RPVB","created_at":"2026-07-05T01:05:27.347053+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PX22RPVBQX5BOAKLCJO5ECX4E6","json":"https://pith.science/pith/PX22RPVBQX5BOAKLCJO5ECX4E6.json","graph_json":"https://pith.science/api/pith-number/PX22RPVBQX5BOAKLCJO5ECX4E6/graph.json","events_json":"https://pith.science/api/pith-number/PX22RPVBQX5BOAKLCJO5ECX4E6/events.json","paper":"https://pith.science/paper/PX22RPVB"},"agent_actions":{"view_html":"https://pith.science/pith/PX22RPVBQX5BOAKLCJO5ECX4E6","download_json":"https://pith.science/pith/PX22RPVBQX5BOAKLCJO5ECX4E6.json","view_paper":"https://pith.science/paper/PX22RPVB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.03687&json=true","fetch_graph":"https://pith.science/api/pith-number/PX22RPVBQX5BOAKLCJO5ECX4E6/graph.json","fetch_events":"https://pith.science/api/pith-number/PX22RPVBQX5BOAKLCJO5ECX4E6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PX22RPVBQX5BOAKLCJO5ECX4E6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PX22RPVBQX5BOAKLCJO5ECX4E6/action/storage_attestation","attest_author":"https://pith.science/pith/PX22RPVBQX5BOAKLCJO5ECX4E6/action/author_attestation","sign_citation":"https://pith.science/pith/PX22RPVBQX5BOAKLCJO5ECX4E6/action/citation_signature","submit_replication":"https://pith.science/pith/PX22RPVBQX5BOAKLCJO5ECX4E6/action/replication_record"}},"created_at":"2026-07-05T01:05:27.347053+00:00","updated_at":"2026-07-05T01:05:27.347053+00:00"}